Divergent and Regioselective Synthesis of 1,2,4- and 1,2,5-Trisubstituted Imidazoles Bruno Delest,†,‡ Prosper Nshimyumukiza,†,‡ Olivier Fasbender,† Bernard Tinant,† Jacqueline Marchand-Brynaert,† Francis Darro,*,‡ and Raphae¨l Robiette*,† Chemistry Department, UniVersite´ catholique de LouVain, Place Louis Pasteur 1, B-1348 LouVain-la-NeuVe, Belgium, and Unibioscreen SA, 40 aVenue Joseph Wibran, B-1070 Brussels, Belgium
[email protected];
[email protected] ReceiVed June 13, 2008
A divergent and regioselective synthesis of 1,2,4- and 1,2,5-trisubstituted imidazoles from a readily available (two steps) common intermediate has been developed. This methodology is based on the regiocontrolled N-alkylation of 1-(N,N-dimethylsulfamoyl)-5-iodo-2-phenylthio-1H-imidazole (10). When this intermediate is engaged in reaction with methyl triflate, selective formation of the corresponding 1,2,5-trisubsituted 1H-imidazole is observed. NMR studies have revealed that this regioselectivity can be accounted for by in situ rapid isomerization of 10 into its 1,2,4-isomer (13) followed by regiospecific N-alkylation of the latter. Conversely, when key intermediate 10 is slowly added to Meerwein’s salt, isomerization can be constrained and regiospecific N-alkylation of 10 leads to 1,2,4-trisubstituted 1Himidazole with a high selectivity. The general character of this methodology has been illustrated by showing that iodine in position 4 or 5 could be easily substituted by an aryl group by Suzuki coupling, whereas the phenylthio group at position 2 could, after oxidation into sulfone, be displaced by nucleophilic substitution.
Introduction Substituted imidazoles are key substructures present in many compounds possessing interesting pharmacological properties.1,2 As a result, a number of strategies have been explored for the †
Universite´ catholique de Louvain Unibioscreen SA. (1) For some reviews on imidazole synthesis, see: (a) Grimmett, M. R. Imidazole and Benzimidazole Synthesis; Academic Press, Inc.: San Diego, CA, 1997. (b) Grimmett, M. R. In ComprehensiVe Heterocyclic Chemistry III; Katritzky, A. R., Ramsden, C. A., Scriven, E. F. V., Taylor, R. J. K., Eds.; Pergamon Press: Oxford, 2008; Vol. 4, pp 143-364. (c) Kamijo, S.; Yamamoto, Y. Chem. Asian J. 2007, 2, 568–578. (d) Du, H.; He, Y.; Rasapalli, S.; Lovely, C. J. Synlett 2006, 7, 965–992. (e) Bellina, F.; Cauteruccio, S.; Rossi, R. Tetrahedron 2007, 63, 4571–4624. (f) Begtrup, M. Bull. Soc. Chim. Belg. 1988, 97, 573–597. (2) For reviews on biological activities of imidazoles, see: (a) Bolani, M.; Gonzalez, M. Mini-ReV. Med. Chem. 2005, 5, 409–424. (b) Jin, Z. Nat. Prod. Rep. 2006, 23, 464–496. ‡
6816 J. Org. Chem. 2008, 73, 6816–6823
preparation of these motifs. One such route consists of the introduction of substituents prior to imidazole ring formation. Suitable precursors may, however, be difficult, or impossible, to synthesize. Many substituents are also sensitive and do not tolerate the reaction conditions for the cyclization. A more general strategy is to introduce substituents to the preformed ring. However, the limitation of such an approach is often a lack of regioselectivity. We are interested in developing a short and divergent synthesis of 1,2,4- and 1,2,5-trisubstituted imidazoles. So, our goal is to find a key intermediate which would be readily available (in a few steps), storable, and could lead selectiVely to both families of compounds, with a large range of substituents. The main challenge faced by the regioselective substitution of imidazoles is the differentiation of positions 4 and 5.1 An elegant way to solve this issue is the introduction of a protecting 10.1021/jo801256b CCC: $40.75 2008 American Chemical Society Published on Web 08/06/2008
Synthesis of 1,2,4- and 1,2,5-Trisubstituted Imidazoles SCHEME 1. Common Strategy for the Synthesis of Substituted Imidazoles
group on the nitrogen atom (position 1) which will direct the substitution at position 5 (or 4). However, when this assistant group has served its purpose and one needs to change it for another, the deprotection/substitution process leads to mixtures of 1,4- and 1,5-isomers (Scheme 1). Indeed, the 5-substituted 1H-imidazole obtained after N-deprotection undergoes a rapid tautomeric equilibrium, and alkylation leads to mixtures in which the 1,4-isomer is usually the major product (due to steric factors).3,4 We reasoned that one way to overcome this problem would be to find a protecting group which could direct both the C(5)substitution and the subsequent N-alkylation. To serve these purposes, we envisaged the use of the N,N-dimethylsulfamoyl group. When positioned on an aromatic ring, the N,N-dimethylsulfamoyl group is known to direct lithiation in the ortho position.5 If position 2 of the 1H-imidazole is already substituted, the presence of this sulfamoyl group on N(1) of the imidazole enables thus highly selective deprotonation and substitution in position 5 (see Scheme 1; PG ) SO2NMe2).1a,6 Moreover, N-sulfamoylimidazoles are known to react with alkylating reagents exclusively via their nonsubstituted nitrogen atom (position 3).7 Alkylation on N(1) is indeed blocked by the N,N-dimethylsulfamoyl group. Formation of a salt by N(3)alkylation increases the lability of the dimethylsulfamoyl group which is then easily removed upon addition of a nucleophile (Scheme 2).7 The outcome of this procedure is thus the N-alkylation of the imidazole ring selectively at the previously nonsubstituted nitrogen atom (numbered 3 in Scheme 2).8 The use of the N,N-dimethylsulfamoyl group then should allow regioselective lithiation/substitution of 2-substituted 1H(3) For some recent examples, see: (a) Lovely, C. J.; Du, H.; Sivappa, R.; Bhandari, M. R.; He, Y.; Dias, H. V. R. J. Org. Chem. 2007, 72, 3741–3749. (b) He, Y.; Chen, Y.; Du, H.; Schmid, L. A.; Lovely, C. J. Tetrahedron Lett. 2004, 45, 5529–5532. (4) For a personal example, see Supporting Information. (5) (a) MacNeil, S. L.; Familoni, O. B.; Snieckus, V. J. Org. Chem. 2001, 66, 3662–3670. (b) Watanabe, H.; Schwarz, R. A.; Hauser, C. R.; Lewis, J.; Slocum, D. W. Can. J. Chem. 1969, 47, 1543–1546. (6) (a) Feldman, K. S.; Skoumbourdis, A. P. Org. Lett. 2005, 7, 929–931. (b) Winter, J.; Re´tey, J. Synthesis 1994, 245–246. (c) Kudzma, L. V.; Turnbull, S. P. Synthesis 1991, 1021–1022. (d) Ngochindo, R. I. J. Chem. Soc., Perkin Trans. 1 1990, 1645–1648. (e) Katritzky, A. R.; Slawinski, J. J.; Brunner, F. J. Chem. Soc., Perkin Trans. 1 1989, 1139–1145. (f) Carpenter, A. J.; Chadwick, D. J. Tetrahedron 1986, 42, 2351–2358. (7) (a) Lee, H. K.; Bang, M.; Pak, C. S. Tetrahedron Lett. 2005, 46, 7139– 7142. (b) Beaudoin, S.; Kinsey, K. E.; Burns, J. F. J. Org. Chem. 2003, 68, 115–119. (8) For punctual examples of 1,5-substituted imidazole synthesis using a similar strategy with other protecting groups, see: (a) Chandana, P.; Nayyar, A.; Jain, R. Synth. Commun. 2003, 33, 2925–2933. (b) Panosyan, F. B.; Still, I. W. J. Can. J. Chem. 2001, 79, 1110–1114. (c) Chivikas, C. J.; Hodges, J. C. J. Org. Chem. 1987, 52, 3591–3594.
SCHEME 2. Alkylation/Deprotection of N-Sulfamoylimidazoles
imidazoles at position 5 (1 f 2; Scheme 3). Subsequent selective alkylation of the nonsubstituted nitrogen atom (2 f 3) and deprotection should then lead to 1,2,4-trisubstituted imidazoles (4). Finally, it has been shown that 5-substituted N-sulfamoyl1H-imidazoles (2) isomerize under equilibrating conditions to give the corresponding, more stable, 4-substituted N-sulfamoylimidazoles (5).9 Application of the alkylation/deprotection procedure (vide supra) to this isomer should allow now obtaining 1,2,5-trisubstituted imidazoles (7) (Scheme 3). Therefore, if R1 and R3 are functional groups which can be easily transformed into a wide variety of substituents, 2,5substituted 1-(N,N-dimethylsulfamoyl)-1H-imidazoles 2 could constitute a general, selective, and divergent route to 1,2,4- and 1,2,5-substituted imidazoles. We selected phenylsulfonyl (R1) and iodine (R3) as transformable functional groups so that our designed key intermediate is 12 (Scheme 4). Indeed, the phenylsulfonyl group at position 2 should be easily displaced by nucleophilic substitution,10 and the iodine atom at position 4 or 5 will enable metal-catalyzed coupling.11 Results and Discussion Synthesis of the Key Intermediate. The synthesis of the intermediate (12) leading to 1,2,4- and 1,2,5-trisubstituted imidazoles was envisaged in three steps. It has been shown that 1-(N,N-dimethylsulfamoyl)-1Himidazole (8) can be monolithiated selectively in position 2.6 Thus, selective deprotonation of 1-(N,N-dimethylsulfamoyl)1H-imidazole (8) followed by addition of phenyl disulfide gives 2-substituted N-sulfamoyl-1H-imidazole 9 in excellent yield (Scheme 4).6a Subsequent oxidation of the sulfide leads to the corresponding 2-phenylsulfonyl-substituted 1H-imidazole 11.12 This latter showed, however, to be poorly stable, leading to (9) Bhagavatula, L.; Premchandran, R. H.; Plata, D. J.; King, S. A.; Morton, H. E. Heterocycles 2000, 53, 729–732. (10) (a) Jones, T. K.; Mani, N. US 2005250948, 2005; Chem. Abstr. 2005, 143, 460319. (b) Bogenstaetter, M.; Carruthers, N. I.; Lovenberg, T. W.; Ly, K. S.; Jablonowski, J. A. WO 2002079168, 2002; Chem. Abstr. 2002, 137, 279192. (11) For some examples of metal-catalyzed coupling involving 4- or 5-iodoimidazoles, see: ref 1c. (a) Yang, X.; Knochel, P. Chem. Commun. 2006, 2170–2172. (b) Langhammer, I.; Erker, T. Heterocycles 2005, 65, 1975–1984. (c) Yoshida, T.; Nishiyachi, M.; Nakashima, N.; Murase, M.; Kotani, E. Chem. Pharm. Bull. 2003, 51, 209–214. (d) Lovely, C. J.; Du, H.; Dias, H. V. R. Org. Lett. 2001, 3, 1319–1322. (e) Kawasaki, I.; Katsuma, H.; Nakayama, Y.; Yamashita, M.; Ohta, S. Heterocycles 1998, 48, 1887–1901. (f) Carver, D. S.; Lindell, S. D.; Saville-Stones, E. A. Tetrahedron 1997, 53, 14481–14496. (g) Cliff, M. D.; Pyne, S. G. J. Org. Chem. 1995, 60, 2378–2383. (12) (a) Moreno, P.; Heras, M.; Maestro, M.; Villagordo, J. M. Synthesis 2002, 18, 2691–2700. (b) Schickaneder, H.; Engler, H.; Szelenyi, I. J. Med. Chem. 1987, 30, 547–551.
J. Org. Chem. Vol. 73, No. 17, 2008 6817
Delest et al. SCHEME 3.
Designed Strategy to 1,2,4- and 1,2,5-Trisubstituted Imidazoles
SCHEME 4.
Synthesis of the Key Intermediate
N-deprotected 2-phenylsulfonylimidazole upon workup conditions. This observation can be attributed to the electronwithdrawing character of the phenylsulfonyl substituent which increases the lability of the sulfamoyl group. In order to keep a high stability of our key intermediate, we decided therefore to redesign our strategy and leave the group at position 2 as a sulfide. The oxidation could indeed be performed later in the synthesis (right before the nucleophilic substitution), giving a new key intermediate 10. This new intermediate is obtained in excellent yield by regioselective deprotonation of 9 in position 51a,6 and iodination. Intermediate 10 is a nice solid obtained in crystalline form which can be stored in the refrigerator for months13 and of which structure could be confirmed by X-ray analysis (see Supporting Information).14 Having our key intermediate in hand (two steps, 83% overall yield), we investigated its potential as a platform for the selective synthesis of 1,2,4- and 1,2,5-trisubstituted imidazoles. Isomerization of 10 into 13. According to our strategy, the access to 1,2,5-trisubstituted 1H-imidazoles from our key (13) Neither isomerization nor degradation was observed after 4 months at18 °C. (14) For both structures determined by X-ray diffraction analysis in this paper, 10 and 18, two independent molecules are observed in the asymmetric parts of the unit cell; they differ only by a slight different orientation of the phenyl group. The bond lengths indicate clearly a greater conjugation in the imidazole ring of 10 as compared to 18.
6818 J. Org. Chem. Vol. 73, No. 17, 2008
SCHEME 5.
Isomerization of 10
SCHEME 6.
Alkylation/Deprotection of 13
intermediate 10 necessitates the isomerization of the latter into its 1,2,4-regioisomer 13 (Scheme 5). As previously described on similar substrates, this can be done at 70 °C in the presence of a catalytic amount of N,N-dimethylsulfamoyl chloride.9 Upon these conditions, the more stable 1,2,4-isomer 13 was obtained in excellent yield (96%). Isomerization could be observed by the low field shift of HC(imidazole) signal from 7.03 to 7.49 ppm in 1H NMR (see Experimental Section for the full characterization data). We have thus shown that both 1,2,4- and 1,2,5-isomers of our key intermediates, respectively, 13 and 10, are readily available in high yields. We investigated then their respective regioselective N-alkylation. Alkylation/Deprotection of 13: Synthesis of 1,2,5Trisubstituted Imidazoles. Previous studies have shown that N-sulfamoyl-1H-imidazoles react with alkylating reagent exclusively via N(3), alkylation on N(1) being blocked by the N-protecting group.7 Accordingly, addition of methyltriflate to 13 gives selectively the salt 14 (Scheme 6). Slow formation of this salt could be detected in 1H NMR when carrying out the reaction in CD2Cl2: H(4) is low field shifted in 14 (7.71 ppm), as compared to 13 (7.50 ppm), and a new CH3 signal appears at 3.83 ppm. Alkylation of 13 increases the lability of the dimethylsulfamoyl group which can then be removed upon addition of methylbutylamine to yield 1,2,5-trisubstituted imidazole 15 in
Synthesis of 1,2,4- and 1,2,5-Trisubstituted Imidazoles SCHEME 7.
Alkylation/Deprotection of 10
good overall yield (80%) and total regioselectivity (no 1,2,4isomer could be detected in the crude mixture).15 Oxidation of 15 by m-CPBA gives quantitatively sulfone 16, which is now on hand for further functionalization in positions 2 and 5. Alkylation/Deprotection of 10: Synthesis of 1,2,4Trisubstituted Imidazoles. According to our strategy, 1,2,4trisubstitued imidazole derivatives could be accessible by regioselective alkylation/deprotection of 10 (see Scheme 3). Key imidazole (10) was reacted with methyl triflate followed by addition of methylbutylamine under the same reaction conditions as for 13 but to our surprise this led to 15 (71% isolated yield), with only traces of 18 (4%) observed in the crude mixture (Scheme 7).15 In order to investigate formation of 15, we performed the reaction in CD2Cl2 and followed it by 1H NMR. No signal which could be attributed to salt 17 was detected. Instead, formation of salt 14 was observed. This suggests a rapid isomerization (faster than alkylation) of 10 into 13 under the reaction conditions, subsequent alkylation/deprotection of 13 yielding 1,2,5-isomer 15 as observed above (see Scheme 6). In fact, it is interesting to note that access to 15 does not actually necessitate a prior isolation of 13 but that 15 can be directly obtained in good yield (71%) by addition of methyl triflate to 10, the isomerization step taking place in situ. However, to access 1,2,4-imidazoles from 10, one needs to be capable of preventing this in situ rapid isomerization. We tried therefore to identify the causes of this isomerization. First, we demonstrated the importance of alkylating reagent in isomerization by showing that imidazole 10 was stable in CH2Cl2 for 5 days. We thus reasoned that the origin of the isomerization could lie in the low nucleophilicity of triflate counterion (Scheme 8; hypothesis 1). Accordingly, we performed the reaction with alkylating reagents producing nonnucleophilic counterions (Table 1, entries 2-4). As anticipated, the use of dimethyl sulfate and Meerwein’s salt enabled formation of the desired isomer 18. The latter is, however, the minor product with the selectivity still being in favor of imidazole 15. The nucleophilicity of the counterion cannot thereby account for all the observed isomerization of 10. We thought then that three other reasons could explain the isomerization (Schemes 8 and 9). First, it is likely that traces of water are present in the medium and act as nucleophile (Scheme 8; hypothesis 2). Second, isomerization may be catalyzed by traces of acid (Scheme 8; hypothesis 3). Finally, (15) As a comparison, methylation of corresponding N-deprotected imidazole leads to a 57/43 mixture of 1,2,4- and 1,2,5-isomers (see Supporting Information).
SCHEME 8. of 10
Potential Causes of the In Situ Isomerization
it is possible that isomerization is mediated by the salt 17 (and 14) (Scheme 9; hypothesis 4). In order to test hypothesis 2 and 3, we carried out the reaction in the presence of, respectively, molecular sieves and a nonnucleophilic base, 2,6-di-tert-butyl-4-methylpyridine (DTBMP)16 (Table 1, entries 5 and 6). Molecular sieves provided exclusive formation of isomer 15, whereas the presence of 10% of DTBMP in the medium did not lead to any significant change in regioselectivity. Isomerization is thereby not due to traces of water or acid.17 Hypothesis 4 was then investigated. In this scenario, at the initial stage of the reaction (
